Introduction: Nash and Moe (NM)is a commonly used method to determine vertebral rotation. Previous studies have reported correlation between the NM grade and vertebral rotation. Ho, et al and Kuklo, et al have provided mathematical formulae establishing this relationship. We have also reported reasonable correlation between the two. However, each of these studies was limited by small numbers.
Methods: We assessed patients with AIS with thoracic or thoraco‐lumbar curves who were candidates for surgical correction and who had preoperative CT scans on file. All measurements were obtained by two independent observers. The apex vertebra of each curve was assigned a rotation grade via the Nash & Moe technique based on standard standing films. The same vertebra was similarly graded on a scout CT (supine) view. Direct measurement of angular rotation was performed on the axial CT slice containing the apex vertebra, as described by Ho et al.
Results: 83 patients (143 curves) were assessed. The group was 77.1% female, with mean age of 14.4 years. Mean coronal Cobb angle was 46 degrees, and mean CT axial rotation was 19.5 degrees. Paired t‐tests showed no significant interobserver differences for any of the measured variables. There was a significant difference between Nash & Moe grades in standing films compared to the same vertebrae on supine views (p=0.000001). A linear regression did not strongly correlate the Nash & Moe grade as seen on supine CT scout view with direct angular rotation (R = 0.49).
Conclusion: Both Nash & Moe grading and Ho et al's technique of CT measurment were found to have good interobserver reliability. However, a significant difference in NM grading is seen with changes in patient position. The previously reported correlation of NM grading and CT scan rotation was found to be weaker in larger samples. This correlation remained unchanged regardless of patient positioning. Previously described formulae establishing this relationship could not be validated.
Significance: Little information can be obtained from the Nash and Moe measurements and surgeons must be aware of its limitations.